Space-time noise reduction system for use in a vehicle and method of forming same

Abstract

A space-time adaptive beamformer for reducing noise in a vehicle that includes two or more microphones. A first weighting network is used for adjusting the signal characteristics of at least one output of the microphones while at least one delay network is also used for delaying the output in time of at least one output of the microphones. A second weighting network then adjusts the signal characteristics of the output of each of the delay networks and at sum adder works to combine the output of the first weighting network and the second weighting network. Finally, an output of the sum adder is combined with an artificial noise free reference signal to provide a low distortion noise reduced output. By generating a desired signal that acts as an artificial noise free signal reference, adequate noise reduction to be obtained without the distortion created due to processing non-linearity.

We claim: 
     
       1. A space-time adaptive beamformer for reducing noise in a vehicle comprising:
 a plurality of microphones, at least one microphone of the plurality of microphones configured to communicate a microphone output; 
 a first weighting network communicatively connected to the microphone output, the first weighting network configured for adjusting a plurality of signal characteristics of the microphone output; 
 at least one delay network communicatively connected to the microphone output, the at least one delay network configured for delaying the microphone output to generate a time delay output; 
 a second weighting network communicatively connected to the time delay output, the second weighting network configured for adjusting the signal characteristics of the time delay output of the at least one delay network; 
 at least one sum adder communicatively connected to the first weighting network and the second weighting network, the at least one sum adder configured for adding a first weighted output of the first weighting network and a second weighted output of the second weighting network; and 
 wherein an output of the at least one sum adder is combined with an artificial noise free reference signal to provide an error signal used to generate a low distortion noise reduced output. 
 
     
     
       2. A space-time adaptive beamformer as in  claim 1 , wherein the plurality of signal characteristics include both amplitude and phase. 
     
     
       3. A space-time adaptive beamformer as in  claim 1 , wherein the artificial noise free reference signal is produced with a normalized least mean squares (NLMS) algorithm. 
     
     
       4. A space-time adaptive beamformer as in  claim 1 , wherein the beamformer is used for providing enhanced audio to a mobile telephone. 
     
     
       5. A space-time adaptive beamformer as in  claim 1 , wherein the beamformer is used with vehicle telematics. 
     
     
       6. A space-time adaptive beamformer as in  claim 1 , wherein a weighting of at least one output of a microphone of the plurality of microphones is adjusted in at least one predetermined frequency band for optimizing performance. 
     
     
       7. A space-time adaptive beamformer as in  claim 1 , wherein the beamformer is located in a vehicle rearview mirror. 
     
     
       8. A space-time adaptive beamformer as in  claim 1 , wherein the beamformer is located in a vehicle rearview display system. 
     
     
       9. A method for reducing acoustical noise in a vehicle using an adaptive beamformer comprising the steps of:
 providing a plurality of microphones with each one of the plurality of microphones having an output; 
 performing a fast Fourier transform analysis on the output of each of the plurality of microphones for dividing the microphone output in to a plurality of frequency bands; 
 adjusting a complex weight of at least one microphone output for each frequency band for determining a fixed beam signal; 
 utilizing a normalized least mean squares (NLMS) algorithm for each of the plurality of frequency bands for providing a reference signal; 
 providing an adaptive beam signal from the fixed beam signal; 
 subtracting the reference signal from the adaptive beam signal for providing an error signal; 
 utilizing the error signal for further adapting the complex weight of the at least one microphone output for providing a noise corrected signal; and 
 supplying the noise corrected signal to an electrical device used in the vehicle. 
 
     
     
       10. A method for reducing acoustical noise as in  claim 9 , further comprising the steps of:
 adjusting the complex weight in both amplitude and phase of the at least one microphone; 
 generating the reference signal so that environmental noise is removed; and 
 locating the adaptive beamformer in a vehicle rearview system. 
 
     
     
       11. A method for reducing acoustical noise as in  claim 9 , further comprising the step of:
 adjusting the complex weight of the at least one microphone output of the plurality of microphones in at least one predetermined frequency band for reducing low frequency noise created by airflow.